Statistical detection of salient points using the Dual Tree M-band Wavelet Transform

In this paper, we develop a novel method to detect salient points in an image from a multiresolution representation. Our contribution is twofold. Firstly, the multiscale representation results from the Dual Tree Wavelet Transform (DTWT) since it enables a great directional selectivity with a reduced redundancy ratio. The second novelty of our work relies on the reliable outliers statistical tests that we apply to detect salient points from the DTWT coefficients. The experiments show the robustness of the approach to noise.